I wrote a little computer program to count letter frequencies. Don't worry you don't need to know anything about computers! Except of course for Googling, Wikipedia, etc.

So, for example, the following extract

It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring of hope, it was the winter of despair, we had everything before us, we had nothing before us, we were all going direct to Heaven, we were all going direct the other way in short, the period was so far like the present period, that some of its noisiest authorities insisted on its being received, for good or for evil, in the superlative degree of comparison only.

Gave me the following (you can check it by hand if you wish):

Word count: 119

Letter counts:

"e" -> 69, "t" -> 48, "i" -> 45, "o" -> 44, "s" -> 42, "h" -> 28, "a" -> 28, "r" -> 27, "n" -> 22, "w" -> 21, "f" -> 19, "d" -> 14, "g" -> 13, "l" -> 12, "p" -> 10, "c" -> 7, "v" -> 5, "u" -> 5, "m" -> 5, "b" -> 5, "y" -> 4, "k" -> 2

Great! I thought. Just as I expected, there are more E's than other letters - 69 of them to be exact. We all know that "E" is the commonest letter in English.

I set my computer to chug away on the internet, analysing lots of similar texts - all in English. All seemed well until I found the following result:

Word count: 145

Letter counts:

"o" -> 71, "a" -> 70, "t" -> 68, "i" -> 59, "n" -> 54, "s" -> 46, "h" -> 33, "r" -> 31, "u" -> 26, "l" -> 25, "c" -> 23, "d" -> 21, "p" -> 18, "y" -> 18, "m" -> 15, "w" -> 13, "f" -> 13, "b" -> 11, "g" -> 9, "k" -> 4, "q" -> 1, "v" -> 1

"O" is the commonest letter!? Obviously my computer has developed a serious fault - or has it?

Please tell me what's going on.

Hint 1

I have tried to put subtle hints throughout the text of the question. One that is not explicit but hinted at is that I searched for the opening lines of novels written by English authors. In fact I did the computer search under Project Gutenberg (not essential to know but narrows the search space for more than one reason). http://www.gutenberg.org/catalog/world/search

Hint 2

I hinted that this can be solved using Google and Wikipedia. If you use the information that @Gareth McCaughan lays out together with the information from hint 1, and apply a Google search, it won't find long to find the exact passage that I analysed.

  • 3
    $\begingroup$ There are many lipogrammatic texts - identifying the right one seems basically impossible without getting really lucky. I'm not sure how this is supposed to be solved. $\endgroup$
    – Deusovi
    Jul 6 '20 at 3:30
  • 1
    $\begingroup$ @Deusovi - I never said it was going to be easy! I'll add some clues soon. $\endgroup$ Jul 6 '20 at 11:29

An obvious guess would be

that your computer has been counting letters in (some portion of) a deliberately e-less piece of writing like George Perec's "La disparition" (translated into English as "A Void", which is more likely what it's reading than the French original).

I remark that

the non-e letters' frequencies look pretty much like normal English.

If this is right then

I'm not sure whether to say that I "knew already". I knew of the existence of "A Void" (and other texts written under the same constraint, but that's the best known) but obviously I didn't know that its first ?page? had those exact letter-counts or that it was the subject of your question in particular.

Some remarks on the hints:

If this is an e-less novel by an anglophone author available on Project Gutenberg, then it seems like it has to be Gadsby. But of course I'd already checked Gadsby; its first 145 words end in mid-sentence and don't have the right letter counts. The only other English lipogrammatic novels I know of are A Void, which is a translation and not on Gutenberg, and Ella Minnow Pea, which is very recent and not on Gutenberg. But I checked those two anyway. A Void doesn't have the right word counts. Ella Minnow Pea doesn't start off lipogrammatic (it gradually suppresses more and more letters as it goes on). So plainly I'm missing something...

  • $\begingroup$ You're close. However, as you remark, the exact letter counts are unlikely to be the same. I'll add information about the word count to make it unique. (Bar one million monkeys typing for eternity of course) $\endgroup$ Jul 5 '20 at 22:23
  • $\begingroup$ Interesting to note that <rot13> B vf gur zbfg pbzzba yrggre. Abeznyyl tbbq Fpenooyr cynlref cersre pbafbanagf yvxr EGF ohg V fhfcrpg gung vs R vf ireobgra gura ibjryf orpbzr zber vzcbegnag guna hfhny, ohg gung qbrfa'g gryy hf jul B vf orggre guna N </rot13> $\endgroup$
    – happystar
    Jul 6 '20 at 7:59
  • $\begingroup$ Those two are generally pretty close anyway. $\endgroup$
    – Gareth McCaughan
    Jul 6 '20 at 12:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.